An early report this week as I’m taking Friday afternoon off. I had a couple of meetings this week, the first on Monday with Marc, where we discussed the Mapping Metaphor visualisation options, the redevelopment of the Historical Thesaurus website and corpus matters. The second meeting was on Thursday with Wendy, Ellen and Marc to discuss the Mapping Metaphor requirements. I sent a couple of example requirements documents that I’d produced for previous projects so people could get an idea of the sort of document that should probably be created, and we reached a decision about how to proceed in the development of such a document. Marc and Ellen are going to map out the sort of access that different users would need and the kinds of visualisations that might be appropriate for these users. After that we will be able to write a more lower level requirements document, and I offered to write this, when the time comes.
I spent the majority of the rest of the week working on the redevelopment of the Historical Thesaurus website. I’ve set up a new database that will house the Thesaurus data, and I have created the table structure for the main data. I’ve also exported the data from Access into CSV format and next week I’ll write a little script that will upload all of this data into the new database on the server. Marc and I discussed the possibility of normalising the data, for example the data currently has many fields containing date information (first, middle and last dates), but these are not all used for every row. From a database design point of view it would make sense to have a separate date table, with each date for a lexeme taking up one row in this table and there being between one and three date rows per lexeme, depending on whether the lexeme has first, middle and last dates. Although this makes sense from a design point of view, it would mean the online database would be quite different in structure to the Access version, and we have decided to keep the structures identical to make import / export of data more transparent and to keep things simple.
I also developed a new interface for the website, based on the existing HT interface as was agreed at the meeting last week. It’s not the most spectacular of designs, due to it being based on the old site, but it is functional and should work well. The new design can be found here: http://historicalthesaurus.arts.gla.ac.uk/new-design/. I am also going to explore a completely new design without any connection to the old site design, which I will tackle next week.